AI

Stop Losing Money on Cleanup Work by Automating the Parts That Don’t Need You

Earmark Team · May 31, 2026 ·

Cleanup and catch-up work is among the most in-demand services accounting firms can sell, and among the hardest to deliver profitably. That was the starting point for a recent webinar led by Megan Reid, a 15-year accounting veteran who started in Big Four, moved through private industry, and now works on the firm enablement team at Digits.

In the webinar, Megan demonstrated how AI-native accounting tools can transform cleanup engagements from time-intensive projects into scalable service offerings. She built a client file from scratch, imported raw PDF bank statements, and walked through an entire cleanup workflow in real time.

Why cleanup work kills profit margins

“Cleanup is obviously valuable work and it’s hard to scale,” Megan said, framing the core challenge clearly. New clients almost always arrive with some sort of mess to clean up. Maybe you have 18 months of uncategorized transactions or transactions that haven’t been posted from the bank feed. You want to take the engagement, but you know it’s going to be hard to make it profitable.

“We always uncover more skeletons in the closet than we think,” Megan noted during the demonstration. If you’re billing fixed fees, you get squeezed by unpredictable hours. Clients want fast turnarounds. Your teams are leaner. “You’re asked to do more with less,” she said.

“Business owners need that work to be done,” Megan pointed out. But the question is “whether or not your workflow lets you take them profitably.”

Breaking down a traditional cleanup shows where the hours go:

Gathering data
Importing it or connecting feeds
Categorizing tons of transactions
Reconciling accounts
Resolving exceptions
Making adjusting entries
Reviewing everything with your client
Delivering the final report

“In a typical 12-month cleanup or catch-up, you spend the majority of your time categorizing and reconciling transactions,” Megan explained. These tasks are also “the most repetitive, pattern-based parts of the job, which is exactly what AI is good at.”

From blank file to categorized transactions

Megan started her demonstration with a completely blank client file, essentially just an empty ledger. She then showed how to handle a common scenario in which a new client hands over a stack of PDF bank statements with no bank login credentials.

She dragged and dropped the first PDF bank statement directly into Digits. “It is extracting all that data from the bank statement, booking it and categorizing it as well,” Megan explained as the system processed the document.

The AI extracted transactions, identified vendors and customers (called “parties” in Digits), populated company logos and descriptions, attached website links, and categorized each transaction into the appropriate account. Megan noted the system pulls from models trained on “more than 800 trillion dollars’ worth of transactions.”

After uploading statements for June through October, hundreds of transactions flowed in. When processing finished, only 12 were flagged for review. “Instead of manually clearing bank feeds,” Megan said, “come here and look at the exceptions.”

These were transactions that required confirmation. Megan clicked into one from Swift Courier Services. The AI suggested “contractors and consultants.” She confirmed it with one click.

From there, the system natively learned from that categorization. It immediately found two similar transactions and offered to update them together. The exception list dropped from 12 to 8 in seconds.

Bank reconciliation without the manual work

Megan demonstrated three ways to get bank statements into the system for reconciliation. You can connect directly to banks like Mercury, Wells Fargo, Chase, and US Bank, which pull statements automatically via API. You can drag and drop PDF statements anywhere in the product. Or you can use email ingestion, where each client gets a unique email address to forward statements.

She uploaded the June statement by dragging it onto the reconciliation screen. The system read the PDF, extracted every line item, and verified each against the ledger. Megan explained that the system uses “pixel bounding boxes” to match statement entries to ledger entries.

June needed one manual step: adding a beginning balance entry that the system couldn’t infer without a connected bank account. Megan created the entry directly in the reconciliation screen. “Unlike legacy systems, where you may have to have three different tabs open and make changes and then come back and refresh, everything can be done directly in here.”

Then she uploaded July’s statement and navigated away. When she returned, it was done. “The statement was uploaded by me. The auto reconciliation was kicked off by Digits and even finalized by Digits,” she showed in the timeline view.

For larger cleanups, Megan recommended uploading multiple statements at one time. Handle any beginning balances in the first month, then subsequent months often complete automatically.

Review tools that surface what matters

Even with AI handling categorization, accountants still need to review and sign off. “It doesn’t replace the accountant. It just removes that tedious work so that you can focus on those judgment calls,” Megan emphasized.

She demonstrated several review approaches. The general ledger view shows all transactions organized like a trial balance, including assets, liabilities, equity, revenue, and expenses. You can filter by status, amount, source, department, or location. Bulk updates work on hundreds of transactions at once.

Megan said the vendors and customers views are her favorite. They each flag two critical items:

New vendors or customers: Any vendor (or customer) the AI sees for the first time in your selected period
Split categorizations: Vendors (or customers) whose transactions appear in multiple categories

“I just need to have eyes on things it has not seen before,” Megan explained. Even if the AI categorized with high confidence, you have final review and say on how it was categorized..

For transactions needing client input, the collaboration happens in one place. Megan showed how to comment on any transaction: “Hey client, what is this for?” The client receives an email with a link, can respond directly in Digits or reply to the email, and the response appears on the platform. “All the collaboration is centralized in one location,” she said, “instead of you having to manage a ton of emails and download Excel files.”

Delivering professional reports, not data dumps

The final step Megan demonstrated was creating custom reports. While the financials inside Digits update live as transactions flow in, cleanup engagements need a formal deliverable, a static document that locks the numbers in place.

Megan built one on screen. She added a cover page, used AI to draft an executive summary, embedded links to the client’s checklist, and configured the financial statements with period comparisons and trend lines. The system includes “hover to discover” insights that show period-over-period changes and what drove them.

When you need to make adjustments after sending a draft, you create a new version. “Any adjustments you’ve made in Digits will then update directly to this report,” Megan explained. Publishing the final version removes the draft watermark and notifies the client.

The platform tracks everything, including when you created the report, when you published it, when the client viewed it, and all comments from either party. You have a complete record of the deliverable and the conversation around it.

“We’ve done 12 months of cleanup in an hour and a half instead of days,” Megan concluded.

What this means for your firm

The key takeaways from Megan’s demonstration show how cleanup engagements can become profitable:

AI categorizes the vast majority of transactions automatically, flagging only true exceptions
Bank reconciliations can run automatically when you upload statements
The system learns instantly from every correction without rules to build or maintain
Your time shifts to reviewing anomalies, making judgment calls, and delivering polished reports

One practical consideration came up during Q&A. When asked about importing messy QuickBooks Online data, Megan confirmed that direct QBO migration exists but cautioned, “You maybe don’t want the AI to learn off of really messy data. You maybe just want to start fresh.” The system uses imported data for baseline training, so starting clean might make more sense for particularly messy files.

For firms trying to grow, this changes the economics of client acquisition. Every prospect with messy books becomes an opportunity rather than a capacity problem. When you can handle cleanup work profitably, predictably, and consistently, you can say yes to more engagements while maintaining margins.

Watch the full on-demand webinar to see Megan’s complete demonstration from blank file to published financials. If you have cleanup engagements in your pipeline right now, consider what your workflow could look like when the repetitive work is automated.

The Accounting Profession’s Favorite Performance Metrics Are Now Dangerously Misleading

Earmark Team · May 20, 2026 ·

PwC Australia cut partners by 35% and staff by nearly 40% since 2023, yet partner income went up 6%. Meanwhile, the IRS says it just had its “most successful filing season in history” with 25% fewer employees. Fewer people are doing more work than ever. But the accounting profession’s core systems for measuring performance, deciding who to hire, and tracking technology investments were built for a different world.

In a recent episode of The Accounting Podcast, hosts Blake Oliver and David Leary talk about a profession transforming from the inside out. From IRS staffing cuts and Big Four workforce reductions to outdated metrics and licensing bottlenecks, we’re seeing technology race ahead while the infrastructure lags.

Tax Season Success Story (with a Catch)

IRS CEO Frank Bisignano told the Senate Finance Committee that the 2025 filing season was remarkably successful despite the agency losing about a quarter of its staff. The IRS received more than 134 million individual returns, 98% of which were filed electronically. Over 90% of filers got refunds in under 21 days, and the average refund jumped 11% to over $3,400.

The agency credited technology upgrades and AI for the performance boost. Using AI and data analytics to identify underreporting, the IRS sent 500,000 letters that prompted corrections, generating $250 million in additional collections. Enforcement revenue was up 12%, and amended return processing improved from six weeks to just three days. Five noncompliance cases alone brought in $2 billion.

“Just five cases and $2 billion,” Blake noted. “That shows there are some real whales out there when it comes to not paying your taxes.”

But David pointed out an interesting wrinkle. There’s still no confirmed IRS commissioner. Bisignano is serving as CEO without congressional approval, yet Congress seems to have accepted this arrangement with little pushback.

Managing by an Outdated Scorecard

For decades, accounting firms have relied on metrics known as LUMBAR: Leverage, Utilization, Margin, Billing rate, and Realization. These metrics made sense when firms billed by the hour and success meant maximizing billable hours. But as AI compresses work time and firms shift to fixed fees and advisory services, these metrics become misleading.

Douglas Slaybaugh argued in Accounting Today that firms need to track different categories entirely. Instead of hours and billing rates, he suggests measuring:

Value creation, like advisory revenue as a share of total revenue
Automation rates
Redefined leverage, like revenue per employee rather than staff-to-partner ratios
Organizational health, including “regrettable turnover,” or losing people you wanted to keep
Client relationships

Blake was blunt about why traditional metrics fail. “If you go over or under on a job based on a job profitability calculation, which is based on hours, it doesn’t actually change anything in the firm because your staff costs are fixed.” When staff are salaried and clients pay fixed fees, being “over budget” on hours is meaningless. “We get so in the weeds,” he added. “We lose the forest for the trees.”

David pushed further, comparing it to Apple before Steve Jobs returned. The company had separate profit-and-loss statements for every product, optimizing each individually while missing the bigger picture. Jobs collapsed it all into one P&L, recognizing Apple as an ecosystem. “Why do you need all these metrics?” David asked. “Focus on the big picture of your firm.”

The shift is already happening at big firms. Client accounting services is the top growth driver for Top 100 firms for the third straight year, with 85% of firms reporting CAS growth. These services now include cash flow forecasting, budgeting, and strategic finance. That work doesn’t fit hourly billing models, yet many firms still try to manage these engagements with traditional utilization targets.

Licensing Rules as a Talent Bottleneck

Current CPA licensing creates what Jack Castonguay of Hofstra University calls a one-way street: firms can hire accountants and train them in AI, but they can’t easily bring in AI experts and train them in accounting.

“The US licensure model almost forces us to start with accountants and teach them AI skills,” Jack wrote in Bloomberg Tax. “It’s good to have accountants who are well versed in AI, but it would be better to also have AI experts trained in accounting. We should create space for both.”

Jack delivered a sharp observation about recent reforms. “We took away the 150-hour moat around the profession, but ultimately built a wall higher for non-accounting majors seeking to become CPAs.”

Blake agreed strongly. “If you can learn accounting theory on your own and pass the CPA exam, why do we require you to go take all these courses? The CPA exam is supposed to test the knowledge. And if you got the knowledge in another way, why do we care?”

The problem plays out in real life. A viewer shared that, despite having a business degree with an accounting minor, Arizona’s requirements and the need for CPA sign-offs create additional barriers for those with non-traditional backgrounds, such as military service.

There’s some progress. Maryland and Nevada joined roughly 30 states adopting alternative CPA pathways that require a bachelor’s degree, two years of experience, and passing the exam, without the 150-hour rule. But David expressed frustration. “We just got past the 150-hour rule, and we’re going to be on this debate and treadmill now for the next five years.”

Meanwhile, big firms aren’t waiting. Beyond PwC Australia’s dramatic cuts, Deloitte US slashed benefits for non-client-facing staff, halving parental leave from 16 to 8 weeks, cutting PTO by five days, and eliminating the $50,000 adoption and surrogacy benefit.

“What if this is just a way to get people to quit so you don’t have to lay them off from AI later on?” David wondered. The timing makes sense. While 51% of workers said they’d quit over return-to-office mandates in 2025, that number has crashed to just 7% in 2026. Workers are scared, and employers know it.

Betting on AI Without Measuring Results

A Thomson Reuters survey of 1,500 professional services respondents across 27 countries revealed only 18% track AI’s return on investment. Forty-two percent don’t measure at all, and 40% aren’t sure whether they do.

“Pretty much 80% aren’t tracking the return on their AI spend,” David said.

Those who do measure focus on the wrong things. Seventy-seven percent track cost savings, 64% track employee usage, but only 26% track client satisfaction, 23% track revenue growth, and just 17% track new business generation.

“They’re not tracking the correct metrics in their firms,” David noted. “This is not an accounting firm problem. This is professional services.”

The risks of poor AI implementation are real. Deloitte faces investigation in Newfoundland and Labrador after a resident discovered its $1.6 million healthcare report contained AI-generated fake citations. This is at least the third Big Four AI incident.

“They’re selling AI consulting services,” David said, “and then they prove they can’t do it themselves.”

The measurement problem extends beyond AI. Annual recurring revenue (ARR), the metric driving virtually every subscription company’s valuation, has no GAAP definition or standardized calculation. Companies define it however they want. A startup CEO recently made headlines for simply making up ARR numbers.

“If I were in charge of accounting standards, SaaS metrics is the first project I would have FASB do,” Blake said. “It’d be the best thing we could do for tech companies.”

The Path Forward

The accounting profession faces a challenge. The technology works, but the supporting infrastructure hasn’t caught up. Firms still manage by metrics that don’t reflect value creation. Licensing rules block the tech talent firms desperately need. And most organizations aren’t even measuring whether their AI investments pay off.

PwC Australia’s CEO, Kevin Burrowes, put it bluntly: “The future is fewer people doing the same amount or fewer people doing more.” Firms that don’t rebuild their internal systems to match this reality risk falling behind in a rapidly transforming profession.

For the full conversation, including discussions about Representative Ilhan Omar’s accounting disclosure error and more details on all these developments, listen to the complete episode.

Why Your Team Resists Change and the Simple Framework That Fixes It

Earmark Team · May 19, 2026 ·

A client builds an AI-powered dashboard, gets his CPA to validate it, then turns around and asks, “So what value do you bring that I can’t get from this thing?” The CPA doesn’t have a great answer. Services get scaled back.

Meanwhile, an oral surgery practice with four doctors and $8 million in annual revenue is still running QuickBooks Desktop, booking revenue through monthly adjusting journal entries, and entering its entire American Express bill as a single payment each month. They haven’t updated a single process since they founded the business decades ago. Both of these clients exist right now, and they could both be sitting in your pipeline this week.

That’s the change landscape accounting firm leaders navigate today. And if you think the biggest threat is AI or the private equity money flooding into the profession, Marcus and Rachel Dillon say you’re looking at the wrong problem.

In this episode of Who’s Really the BOSS?, the Dillons, owners of Dillon Business Advisors, make the case that the real risk isn’t the change itself. It’s how you lead your people through it. Drawing on real client stories, their own leadership missteps, and a framework borrowed from Patrick Lencioni, they lay out a practical approach to change management any firm leader can start using immediately.

The Change Landscape: From Silicon Valley to Main Street

Before you can bring your team through change, you need to understand what you’re actually up against. The answer depends on where you’re standing.

Marcus spends time networking with partners at top-20 and top-100 firms with $60 million or more in revenue. What he hears from those conversations tends toward doomsday. These firms serve private equity-backed businesses whose principals all have finance or business backgrounds. Those clients are leaning hard into AI, meaning the professionals serving them have to keep pace or move faster.

One leader at a larger firm told Marcus he no longer opens conversations with “How are the kids?” Instead, the first question he asks clients, prospects, and peers is, “How are you using AI today?”

“If your clients are changing faster than you are,” Marcus explains, “you’re going to be the weakest link in that relationship, and they’re going to move on faster than you can.”

The Big Four are already placing their bets. PwC is doubling down on technology and AI at the entry level, slashing recruiting and campus visits. If that layer of the workforce shrinks, they don’t need to wine and dine as many college students. EY is taking a different approach, doubling its CPA exam pass bonus to $10,000 and investing in the human side.

But while Silicon Valley types are sounding the alarm, Main Street tells a different story.

Remember that oral surgery practice? The lead doctor told Marcus they set up the business nearly 30 years ago and never updated their processes because the same team has been in place the whole time.

DBA’s plan for this client is to set up QuickBooks Online, enable bank feeds, connect them to a service like Ramp, and automate the revenue journal entry. Low-hanging fruit by any modern standard.

“You have to choose how analog you want to exist in this digital world,” Marcus says. The clients who want a human touch continue to pay a premium for it. A purely digital product, he argues, is a race to the bottom.

When Change Communication Goes Wrong

Marcus doesn’t sugarcoat DBA’s early track record on change communication. When the firm merged in another practice nearly a decade ago, Marcus was so excited about the acquisition that he gathered everyone in the conference room and essentially announced it cold. Most team members were hearing about it for the first time.

“That probably didn’t go over as well as I could have hoped,” he admits.

The fallout from moments like this is bad. People disengage. The service atmosphere turns mediocre. Tension builds. Marcus found himself labeled “addicted to change,” which bred resistance rather than readiness.

“If you don’t work on your culture, you still have a culture,” he says. “It’s just unintentional. The same can be said of change.”

Rachel offers the perspective from the other side. When she talks to team members about why they push back on change, the answer is almost always a lack of clarity. They don’t understand why it’s important. They can’t see how it impacts them personally.

“A lot of times it feels like, ‘This is going to take me longer and I’m going to have to work more. And I don’t have any more hours or capacity left to give,'” Rachel explains.

The Dillons evolved toward a three-question framework:

What is changing?
What is staying the same?
How does this impact me?

It was an improvement, but still incomplete. It only addressed the team’s perspective, not clients or other stakeholders.

A peer group introduced Marcus to Patrick Lencioni’s Four Ps framework. The Dillons adopted it as their change-management filter and introduced it to the team at their recent Gather event alongside their rally cry for 2026: “Lead change, create impact.”

The Four Ps: Your Repeatable Framework for Leading Change

The framework gives firm leaders four sequential steps to follow every time they introduce change, whether it’s a new tech stack, a team restructuring, or a client exit strategy.

Purpose: What are we changing?

You need to anchor every change in something bigger than “we found a cool new tool.” At DBA, that anchor is their mission, vision, and values. Their core values spell out the word IMPACT, and Rachel describes how they literally map each proposed change back to specific letters in that acronym.

The trap most leaders fall into here is vagueness. Marcus admits he’s guilty of softening language because he wants to be liked and avoiding directness to dodge conflict.

“Just tell me what you expect. Just tell me what you need me to do,” Rachel says. “People don’t want 20 options. They want one or two.”

Marcus borrows from Andy Stanley: “To be clear is to be kind. To be unclear is to be unkind.”

“If you can’t clearly say what’s changing, the team will default to their comfort level,” Marcus warns. “Which means they’ll do as little as possible.”

Picture: What does success look like?

Leaders often skip this step. They explain what’s changing and how it will happen, but they never describe what winning looks like on the other side.

Marcus uses a family vacation analogy. You decide to take a trip (that’s the purpose). Now tell the kids you’re going to Disneyland and describe the destination so everyone can see it.

In a firm context, that might mean showing the team what life looks like after implementing a team-of-three service model: predictable capacity, no more overtime scrambles, better client satisfaction scores.

The Dillons deploy an exercise called Optimist/Pessimist. Pair people up. One person must articulate at least one or two positives about the proposed change. The other must find negatives. This gives explicit permission to voice concerns that would otherwise get whispered in private channels.

“Once we are sick of saying the same thing over and over again, they’ve actually received it, processed it, and can carry it out,” Rachel says.

Plan: How do we get there?

The plan phase breaks the picture into executable steps. Extending the road trip metaphor, explain whether you’re flying or driving. If driving, are you taking the scenic route? Where do you pull over to celebrate progress?

Rachel emphasizes two non-negotiables for every step: a responsible person and a deadline. Each milestone needs an owner and a date, so there’s no ambiguity about who’s doing what by when.

This is also where you appoint change agents from within your team. Team members who showed energy during the Picture phase are natural candidates to lead portions of the execution.

“A simple plan executed beats a perfect plan that’s been delayed,” Marcus notes.

Part: What’s my role in this?

Every single person needs to understand their role, including those whose role is “nothing changes for you.”

Marcus shares a recent example from DBA’s acquisition work. For some team members, the message was, “Keep serving your current clients well. You’re not getting new clients from this acquisition. You’re not learning a new process or technology.”

Simply telling people “your job stays the same” is just as critical as the detailed instructions given to people at the center of the transition.

When you don’t tell people their part, they default to their worst experience. Maybe a previous boss promised “nothing will change” and then changed everything. You can’t control the baggage people carry, but you can replace old narratives with present-tense clarity.

This step requires a conversation, not an email. People need two-way dialogue where they can ask questions and process in real time.

Leading Through the Messy Middle

Marcus closes with an honest confession. “I’m as guilty as anybody. I want to initiate the change. And I want all the fruit from the success of that change. I don’t want to live through the change. I want to just speed through it or delegate it.”

Successful firms have leaders who bring their people through change intentionally, with clarity, conviction, and care.

The Four Ps give you a repeatable filter for any transition:

Purpose: Anchor the change in your mission and values and say it plainly
Picture: Show people what success looks like, then repeat until you’re sick of it
Plan: Break the vision into steps with owners and deadlines
Part: Tell everyone their role in a live conversation, not an email

Whether you’re navigating a firm acquisition, a technology overhaul, or wondering how fast AI is coming for your services, the same four questions apply. As the Dillons put it, the goal for 2026 is to lead change and create impact.

Listen to the full episode to hear Marcus’s take on how fast AI is really moving, Rachel’s breakdown of the Optimist/Pessimist exercise in action, and why moving homes during busy season might actually make perfect sense for a couple “addicted to change.”

Rachel and Marcus Dillon, CPA, own a Texas-based, remote client accounting and advisory services firm, Dillon Business Advisors, with a team of 15 professionals. Their latest organization, Collective by DBA, supports and guides accounting firm owners and leaders with firm resources, education, and operational strategy through community, groups, and one-on-one advisory.

AI Can Reconcile a Bank Account End to End Without Instructions

Earmark Team · May 15, 2026 ·

The accounting profession sees AI companies building tools that weren’t meant for accountants but are increasingly doing accountants’ work. On Episode 482 of The Accounting Podcast, hosts Blake Oliver and David Leary discussed Perplexity and Palantir’s moves into core accounting territory while most firms struggle to see any productivity gains from their AI investments.

Trump Accounts Hit 90% Adoption in First Year

Before getting into AI disruption, the hosts discussed a surprising government success story. Around four million children have been signed up for Trump Accounts in the program’s first year. That’s an 85-90% adoption rate among eligible births since January 2025.

“Think about 401(k) plans,” David pointed out. “It’s been decades and they’re only at 35-40% participation. With college savings plans, it’s been 25 years and 25% participation.” The difference is Trump Accounts combine simplicity (a one-page form filed with your tax return) with immediate value (a $1,000 government contribution, plus additional funds from donors like Michael Dell for low-income families).

Blake ran through the potential impact. If families contribute the maximum $5,000 annually, a child could have $271,000 by age 18 based on historical S&P 500 returns. “That’d be pretty nice,” he said. “Turn 18 and you get $271,000. Maybe that would be a down payment on a house someday.”

Perplexity and Palantir Target Core Accounting Work

The real disruption story started with a simple Instagram ad that caught David’s attention. It read “Find every duplicate entry hiding in your QuickBooks.” The advertiser wasn’t a QuickBooks app developer or accounting software company. It was Perplexity, an AI company better known as a search engine competitor.

Clicking through revealed Perplexity’s new QuickBooks “health check” offering. It includes P&L analysis, expense categorization, AR/AP aging reviews, and reconciliation audits. Perplexity’s homepage also features tax-specific prompts, and they’ve officially launched “Computer for Taxes,” an AI agent that drafts full federal tax returns.

“Perplexity is doing what a whole company just launched to do,” David observed, referring to startups like TaxGPT that built entire businesses around AI tax prep.

Meanwhile, the IRS is testing a $1.8 million pilot with Palantir to build an AI system called SNAP for selecting audit targets. The system will analyze both structured and unstructured data to identify potential fraud and noncompliance.

“People have called Palantir the most dangerous company on earth,” David noted. “So that’s who’s going to pick audits now for the IRS. It’s a little bit scary.” The system could potentially cross-reference tax returns with e-commerce storefronts, social media activity, and data from other government agencies where Palantir operates.

Blake’s Test Whether AI Actually Does the Books

Blake decided to test whether these AI tools could handle real accounting work. He asked Claude Cowork to reconcile a brokerage account in Xero that had no bank feed connection.

Without detailed instructions, Cowork created its own task list, asked clarifying questions, then got to work. It parsed the PDF bank statement, converted it to a CSV formatted for Xero’s import requirements, imported it as a bank feed, matched transactions through the Chrome browser, ran the reconciliation report, and exported it as a PDF.

“End to end, no hand-holding,” Blake said. Even better, he had Cowork save the entire process as a reusable “skill” that improves with feedback. “Now I can upload a PDF and say ‘reconcile this account’ and it will do everything as I like it.”

This raises questions for accounting software companies. “What is the point of Jax in Xero then?” David asked. If external AI agents can perform full reconciliation by interacting through a browser, why build internal AI at all?

Blake says Xero built Jax by copying chatbot functionality from ChatGPT. It’s limited to conversation windows with no ability to handle multi-step workflows. “The AI platform companies have simply raced ahead,” he said.

Firms Are Automating the Wrong Things

Despite all this technological capability, 80% of firms report no measurable AI impact on productivity after three years and billions in spending. The problem, according to a CFO.com opinion piece Blake highlighted, is that firms use the wrong measurement framework.

Most organizations use cost accounting, which rewards local efficiency gains, including hours saved here, a task automated there. But they should use throughput accounting, which asks whether AI actually removes the main bottleneck limiting revenue generation.

“You may have theoretically saved all these hours on tax prep, but if all the returns pile up at review, it’s not getting to the client any faster,” Blake explained. “You’re not delivering any additional value.”

Both hosts agreed the real bottleneck in most accounting workflows isn’t doing the work; it’s getting the information needed to do it. Client document collection causes the biggest delays.

Blake sketched a solution: an AI agent that monitors client folders, compares submissions against a requirements list, and automatically follows up on missing or incomplete documents. “AI won’t be tired on day two or day three,” David said. “It can review these things and send the email again tomorrow. And it’s never going to complain.”

The Three-Person Firm of Tomorrow

The hosts painted a picture of accounting’s near future: firms where a few people do the work of ten, supported by AI agents handling staff-level functions.

“Instead of having 30 clients at your firm, you now can have 90,” David said. “That makes a huge difference.”

Blake envisions a new role structure that includes a CPA or tax expert, a dedicated technologist, and an apprentice, plus “half a dozen AI agents doing the staff functions.” He pointed to conversations with Peter McCarroll of Fuel Accountants, who predicts every firm will need a technologist role.

But Blake offered a reality check. “Cowork crashes. Agents are slow. Building a reliable tech stack in a period of this much change is genuinely hard.” Firms don’t need to adopt every new tool. But they do need to understand their workflows well enough to know exactly where to point AI.

General-purpose AI platforms are building accounting capabilities faster than the profession anticipated. They’re not waiting for permission or partnerships. Listen to the full episode for more insight into figuring out where your real bottlenecks are and pointing AI at them, or watch as others (maybe even your clients) use these tools to work around you.

The AI Trust Problem Accounting Can’t Afford to Ignore

Earmark Team · May 15, 2026 ·

Here’s a thought experiment. You ask ChatGPT to write you a research summary. It comes back 70% accurate. You tweak a few paragraphs, fix some facts, and you’re good to go. Now imagine that same 70% accuracy rate applied to your general ledger, audit workpapers, or financial statements.

As Mark Hickman, Sage’s Managing Director for North America, puts it bluntly, “You go to jail for that.”

That line from Episode 34 of The Unofficial Sage Intacct Podcast cuts straight to a tension every CFO, controller, and accounting professional grappling with AI needs to understand. The technology promising the biggest efficiency gains in a generation operates in a profession where approximate answers are a liability.

Hosts Doug Lewis, Matt Lescault, and Emily Madere sat down with Mark for their second annual Sage Future conference preview episode. Mark oversees Sage’s largest and fastest-growing region. He’s watched the Intacct acquisition grow more than 10x over the past 11 years. He’s in front of customers and partners constantly.

When he talks about what finance leaders say about AI, it’s field intelligence from someone who’s been in tech for nearly 25 years and has seen every major shift from dial-up internet to cloud computing.

He argues that while every corner of technology races to bolt on AI capabilities, accounting demands a fundamentally different approach built on trust, traceability, and human control. And the companies best positioned to deliver that are the established platforms sitting on vast reservoirs of trusted data.

Accounting AI Can’t Be a Black Box

Strip away all the marketing noise around artificial intelligence and you land on a simple question: where did that number come from?

In most industries, that question is nice-to-have. In accounting, it’s everything. And it’s why Mark frames Sage’s entire AI strategy around three pillars: trust, control, and accountability.

“When you type into ChatGPT, write me an essay on this or write me a paper on this, it can be 70% right and you can tweak it,” Mark explains. “You can’t be 70% right in accounting.”

Trust means AI outputs can’t disappear into a black box. Sage built what Mark calls a “trust label.” It’s a mechanism that lets users click into any AI-generated output and trace exactly where the data came from and how AI reached the conclusion. Think of it as an audit trail for the AI itself.
Control means humans stay in charge. “Accounting is different,” Mark emphasizes. “We can’t just have AI running everything behind the scenes. We need humans to control that AI and deliver what they need from those outputs.” The AI changes day-to-day workflows, but it assists rather than drives.
Accountability means everything is traceable. “We don’t want some large language model that somebody’s just pumping stuff into. It’s coming back. You have no idea where it came from, how that data was trained. Is it hallucinating? Is it not hallucinating? You need to be able to trust that that AI is credible, and you’re going to be able to use it in your accounting when you produce that to the auditors.”

Emily pushed Mark on a question many Intacct users ask: what about these new solutions flooding the market that claim to be “AI first”?

Mark’s response was diplomatic but pointed. New organizations saying “we’re AI native” and driving innovation are “good things for our industry.” But “how do you train AI? You train AI with data. We have a lot of data.”

Beyond raw data, Sage just kicked off what Mark calls its “Agentic AI marketplace.” This is a framework where partners build specialized AI agents that work across the broader Sage ecosystem. “We’re building hundreds and hundreds of agents that will be available to our customers,” he notes. The company is taking a platform approach where domain expertise gets layered onto trusted financial infrastructure.

The Adoption Paradox: Faster Than Cloud, Slower Than the Hype

Mark brings perspective from his 25 years in tech. He remembers when the internet arrived on dial-up connections that took five minutes to load. He watched the cloud evolve from radical concept to default infrastructure. Now he’s seeing AI reshape everything again.

“People thought it was going to move much quicker than it actually is,” he observes. “Adoption in these things is way more complicated than actually delivering the tech for it.”

In each major tech shift, people overestimate adoption speed. “The cloud is still being adopted in some places,” Mark points out.

The hosts brought up a perfect example of the hype-reality disconnect. Allbirds, a shoe company, rebranded itself as an AI organization and watched its stock rocket 600% in a single trading session before promptly crashing. “It’s reminiscent a little bit of the dot-com boom,” Mark says, “where people had a website and therefore their business was worth billions. And then everybody figured out they didn’t actually have a business case.”

But he distinguishes this moment from that bubble. “If you look at AI, it’s being driven predominantly by very large companies that are established with lots of customers and lots of money.” The recent tech stock dip “is just a reset on the adoption.”

So what are finance leaders actually saying when Mark sits across from them?

“We’re hearing this is game changing,” he reports. But it’s game changing in specific, practical ways:

Efficiency that enables strategy. “They look at AI to help them be more efficient so they can be more strategic,” Mark explains. The role of the CFO is becoming “significantly more strategic.”
Faster closes. “We’ve been talking for years about reducing time to close and eliminating the month-end close. AI is really going to speed that up.”
Immediate productivity gains. “When we let customers use our agents, they’re like, ‘I’m three times more productive. I cannot believe how much faster we’re getting things done.'”

But adoption takes time. “You can’t just inject things like that into your business overnight,” Mark cautions. “It’s got to be done in a way that makes sense to your workflows and your teams and how your processes roll.”

The conservative pace of AI adoption in finance is an essential feature in a profession where errors carry legal consequences.

Beyond the AI Headlines: What Else Is Changing

Matt asked for more insight into some other things happening at Sage that might not get as much attention because of how much focus is on AI.

Mark shared several initiatives that may have immediate impact:

Speed to market and faster implementations. “How are we going to implement faster? How are we going to get customers’ time to value reduced?” Mark asks. Matt reinforced why this matters. “If we have implementations that take three, six, eight months, we’re going to lose on that side of things.”
Vertical and micro-vertical specialization. “Our solution addresses the needs of those businesses within those verticals, which is something we’ve always done, but we’re doubling down on that,” Mark explains.
Strategic acquisitions filling gaps.
- Expense management. “We’re about 12 months into it and it’s done incredibly well. Overachieved all targets. Customers love it.”
- Sage HCM (payroll and HR). “It’s a great technology that’s really going to help us grow our business and deliver for our customers in a complete solution.”

Emily confirmed these acquisitions address real needs. “That was a big gap. In the past, a lot of clients asked for an expense management solution and wanted it all in one.”

What to Expect at Sage Future 2026

The Sage Future conference runs April 28-30 in San Francisco. Based on last year’s feedback, Sage restructured the entire event.

The new three-tier structure includes:

Keynotes now feature primarily external voices. Mark will interview Scott Krug, CFO of the New York Yankees. “He closes the books and does audits just like everybody else,” Mark says. Kara Swisher will discuss AI trends.
Super Sessions are new, product-specific deep dives.
Breakout sessions provide the next level of detail for features that caught your attention.

Matt made an important observation: “I’ve seen a lot of clients implement Intacct and then don’t go back to reconfigure over their lifetime. They’re not getting the full feature set out of the product.” The Super Sessions directly address this education gap.

Other highlights include:

Three embargoed announcements that Mark calls “very exciting and game changing for customers and partners”
CPA.com as titanium sponsor, validating Sage’s positioning as “accountants who serve accountants”
Monday night at Oracle Park (Giants stadium) and Thursday’s Sage Fest at an undisclosed “top” San Francisco venue
Post-event content distribution through webinars for those who can’t attend

“We want people to leave understanding where we’re going as a business and how we’re supporting them,” Mark says.

The Question Every Finance Leader Should Ask

The thread running through Mark’s conversation reframes how accounting professionals should evaluate every AI pitch landing on their desk.

A 70% accuracy threshold works for drafting marketing copy, but it’s a non-starter when the output feeds financial statements and regulatory filings. That constraint reshapes everything about how we build and deploy AI in accounting.

Whether you’re a CFO weighing technology investments, a controller separating AI substance from hype, or a partner building around the Sage ecosystem, listen to the full discussion. And if you’re heading to Sage Future 2026 in San Francisco, you now know exactly what to look for.